Creating art through artificial intelligence

Generai creates artworks based on its users’ imaginations. As explained in a press release, this platform was developed in the Student Project House at the Swiss Federal Institute of Technology in Zurich.

The four students have used new types of model called diffusion models, which are still being researched. “We’re kind of at the front of it,” one student explains. “Not only our own models but trying to do our own research so that we bring the newest available technology to the broad masses.”

Users simply enter text describing the work of art they want to create and then choose from suggested keywords and a specific style of painting. “It will then be rendered on the best computers we can rent at Google.” Generai then delivers four unique images in high resolution that can be downloaded. To take advantage of the full range of possibilities, users buy credits for 10, 25 or 75 creations for between US$9.99 and US$44.99. Users can also order a printed version of the images on metal, canvas or as a poster.

The students emphasize that their goal is not to replace human artists through this technology. Rather, they want to enable them with a tool that “especially artists can use for whatever they like, whether it be background or concept art or for imagination”.